Oppdateres daglig1507 bekreftede historier
Kildesjekket AI-dekning av produktlanseringer, endringer i retningslinjene, sikkerhetsundersøkelser og bransjebevegelser, forklart på vanlig engelsk av et nonprofit-utdanningsteam.
Hver historie kobler til de sterkeste tilgjengelige bevisene: originale kilder når de er tilgjengelige, ellers tydelig tilskrevet rapportering.
Hva skjedde, hvorfor det er viktig og hva du bør se på – ingen sjargongskatt.
Når signalet er tynt, publiserer vi ingenting i stedet for å polstre feeden.
En voksende strøm av bekreftede perspektiver for folk som trenger å forstå AI uten å jage etter hype.
The Motley Fool reports that AMD’s next-generation Instinct MI455X accelerator is specified with 432GB of HBM4 memory, 50% more than the MI355X. The outlet also reports that AMD’s Helios rack system is in production, with initial shipments expected later this quarter.
NAI500 reports that opposition to AI data centers over electricity, water and local development is entering Wall Street risk analysis and state-level politics ahead of the 2026 U.S. midterm elections.
TechNode reports that Chinese AI-chip developer Enflame Technology will begin online and offline subscriptions for its Shanghai STAR Market IPO on Sept. 2, seeking RMB6 billion to develop and commercialize its fifth- and sixth-generation chips and related hardware-software systems.
A new arXiv paper presents an open-source policy schema and supporting tools intended to let organizations define content rules for generative AI applications and apply them across training, alignment and runtime monitoring.
Qwen describes Qwen3.8-Flash-Next as a multimodal mixture-of-experts model with 125 billion total tokens and 6 billion active parameters, and says it previews architecture planned for Qwen4.
A preprint describes LUCAID, an agentic multimodal AI system that integrates nine lung-cancer pathology tasks and reports 93.0% concordance with an expert-panel reference standard in prospective clinical validation.
An arXiv paper introduces DRRG, a discrete-diffusion framework that generates radiology reports through iterative masked-token denoising. The authors report stronger results than comparison methods on MIMIC-CXR and CheXpert Plus across several metrics, while using a smaller language-model decoder.
Blackbird has raised Australia’s largest venture-capital fund, while general partner Samantha Wong says the firm is becoming more cautious about crowded AI applications and expects to invest more in infrastructure, semiconductors and other deep technologies.
A new arXiv paper proposes VisCache, a no-training framework that reduces visual KV-cache storage for vision-language models while retaining 19% to 28% of the cache. The authors report speedups of up to 2.35 times, but the results have not been independently validated here.
A new technical report introduces WeMM-Embedding, a family of 2B, 4B and 9B models for representing text, images, videos and visual documents in a shared space. The report says the models improve internal WeChat benchmarks and have been deployed across search and recommendation services.
A new arXiv study reports that tuning the blend between text and image prototypes is not the main limit on few-shot vision-language model accuracy. In experiments spanning 4,800 evaluation cells, validation-free linear probes outperformed even a blend selected with test-set information.
A new arXiv study comparing primate vision with multiple neural-network designs reports that predictive world models were most robust when object appearance changed, pointing to progressive integration of motion as a missing principle for dynamic AI.
En nyttig orientering hver uke
Få ukens bekreftede AI-nyheter, originale data, nyttige verktøy, læringsvalg og ferske AI-jobber.
Ansette en AI-profesjonell eller lansere et nyttig AI-produkt? Sett det foran folk som kom hit for å lære og handle.
Legg ut en AI-jobb Send inn et AI-verktøy