Live Stream from the Namib Desert https://bookofjoe2.blogspot.com/2025/10/live-stream-from-namib-desert.html
Blogspot
Live Stream From the Namib Desert
4Chan Lawyer publishes Ofcom correspondence https://alecmuffett.com/article/117792
Dropsafe
The Ofcom Tea Party: 4Chan Lawyer publishes Ofcom correspondence, British regulator claims “sovereign immunity” to defend itself…
Ofcom — driven by the letter of British law that they are bound to follow, but still Ofcom — are quietly making Britain look (a) very silly and (b) as if we haven’t yet shucked-off the Americ…
Andrej Karpathy – It will take a decade to work through the issues with agents https://www.dwarkesh.com/p/andrej-karpathy
Dwarkesh
Andrej Karpathy — AGI is still a decade away
"The problems are tractable, but they're still difficult”
Claude Skills are awesome, maybe a bigger deal than MCP https://simonwillison.net/2025/Oct/16/claude-skills/
Simon Willison’s Weblog
Claude Skills are awesome, maybe a bigger deal than MCP
Anthropic this morning introduced Claude Skills, a new pattern for making new abilities available to their models: Claude can now use Skills to improve how it performs specific tasks. Skills …
AMD's Chiplet APU: An Overview of Strix Halo https://chipsandcheese.com/p/amds-chiplet-apu-an-overview-of-strix
Chipsandcheese
AMD’s Chiplet APU: An Overview of Strix Halo
Hello you fine Internet folks!
Root System Drawings https://images.wur.nl/digital/collection/coll13/search
AWS multiple services outage in us-east-1 https://health.aws.amazon.com/health/status?ts=20251020
Alibaba Cloud says it cut Nvidia AI GPU use by 82% with new pooling system https://www.tomshardware.com/tech-industry/semiconductors/alibaba-says-new-pooling-system-cut-nvidia-gpu-use-by-82-percent
Tom's Hardware
Alibaba Cloud says it cut Nvidia AI GPU use by 82% with new pooling system— up to 9x increase in output lets 213 GPUs perform like…
A paper presented at SOSP 2025 details how token-level scheduling helped one GPU serve multiple LLMs, reducing demand from 1,192 to 213 H20s.
BERT is just a single text diffusion step https://nathan.rs/posts/roberta-diffusion/
nathan.rs
BERT is just a Single Text Diffusion Step
A while back, Google DeepMind unveiled Gemini Diffusion, an experimental language model that generates text using diffusion. Unlike traditional GPT-style models that generate one word at a time, Gemini Diffusion creates whole blocks of text by refining random…
Wikipedia says traffic is falling due to AI search summaries and social video https://techcrunch.com/2025/10/18/wikipedia-says-traffic-is-falling-due-to-ai-search-summaries-and-social-video/
TechCrunch
Wikipedia says traffic is falling due to AI search summaries and social video | TechCrunch
Looks like Wikipedia isn't immune to broader online trends, with human page views falling 8% year-over-year, according to a new blog post from Marshall Miller of the Wikimedia Foundation.
60k kids have avoided peanut allergies due to 2015 advice, study finds https://www.cbsnews.com/news/peanut-allergies-60000-kids-avoided-2015-advice/
CBS News
Some 60,000 kids have avoided peanut allergies due to landmark 2015 advice, study finds
About 60,000 children have avoided developing peanut allergies after 2015 guidance upended medical practice by recommending introducing the allergen to infants starting as early as 4 months, a new study finds.
KDE Connect: Enabling communication between all your devices https://community.kde.org/KDEConnect
Neural audio codecs: how to get audio into LLMs https://kyutai.org/next/codec-explainer
kyutai.org
Neural audio codecs: how to get audio into LLMs
Why modeling audio is harder than text, and how to make it feasible with neural audio codecs.