These are the use cases for AI I’m extremely excited and happy about. Somebody had a question but not the means to answer it. So, he used an AI to do the grunt work and found an amazing answer that he would never have found without these insanely powerful tools at his disposal. Absolutely brilliant.

mr. TIM@timkellogg.me · 4d ago

a guy found a planet 117 light years away using Claude Code with Opus 5.5 & Fable NASA gave him telescope time to acquire better proof The agents used traditional astronomy tools for processing, search, transit fitting, and false positive checks www.reddit.com/r/ClaudeAI/c...

The first library I scanned with Mythos was Phoenix and it found: Nothing! This is an incredible statement for the quality of the library and the thoroughness of the engineering behind it. Absolutely outstanding work, dear Phoenix team! #ElixirLang

Peter Ullrich@peterullrich.com · 4d ago

I now have access to Mythos 5.1 RIP our CIs

Dwight Schrute Salutes in The Office

Alt: Dwight Schrute Salutes in The Office

static.klipy.com

bryan newbold@bnewbold.net · 5d ago

ah damn, RIP Margaret Hamilton. she took safety-critical software engineering seriously, and got people to the moon and back. news.mit.edu/2026/margare...

No sabía q en 2003 Disneyland Paris pidió a la familia de Luis Buñuel poner una calle a su nombre y el hijo les contestó q ok pero q quería 10 millones de € y q la calle se llamara «Boulevard Luis Buñuel-Me cago en Dios y el fascista de Walt Disney». Pero qué puta maravilla es esta. Os dejo la carta

I have absolutely no idea how anyone or any agency could ban recursive self-improvement. Alternately, I suggest you legislate a ban against naked pointers in programming languages, for at least that would gain you some political capital among conservatives.

Congressman Greg Casar@repcasar.bsky.social · 3w ago

You may have heard about my AI bill with @sanders.senate.gov. We're introducing the legislation today, and I wanted to highlight one of the most important aspects: A ban on “recursive self-improvement.” What does that mean? 👇🏽

"I have nothing to hide." False. You have medical history You have financial behavior You have location patterns You have political views You have relationship data Privacy was never about hiding crimes. It was always about owning your life.

"You don't have to review the code, it's good". The code:

"Nunca se sabe de qué peor suerte te ha salvado tu mala suerte".  (Cormac McCarthy)

Because pelicans on bikes are commonly tested I run a modified benchmark with Andalites which not only tests visual quality but also common sense/ intent Astra is the first model to understand it needs a tandem design bsky.app/profile/snep...

max power 🌌@sneptech.bsky.social · last mo.

🅷🅾🅻🆈 🆂🅷🅸🆃

"Compared to nukes, tech shutdowns are amazing. Thanks to Big Tech, America – and only America – can brick almost any country on earth, shutting down its administration, agriculture and industry without firing a single shot."

By Cory Doctorow (GPG 0xBF3D9110957E5F4C)@doctorow.pluralistic.net · last mo.

If you'd like an essay-formatted version of this thread to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog: pluralistic.net/2026/09/03/b... 2/

The Origin of Consciousness (2008) | Discussion

The Origin of Consciousness

From the highly eclectic blog of Mark Dominus

blog.plover.com

Models Are Getting Dumber on Purpose | Discussion

Models Are Getting Dumber on Purpose - Walter van der Giessen

Reasoning scores climb while per-token compute drops. Labs are stripping world knowledge out of models, and I think that's the right trade.

w4g1.dev

Stripe Clinches over $7B Deal to Buy AI Firm OpenRouter | Discussion

This blog post by @whatyouhi.de is amazing. The best explanation of GenStage I've seen. andrealeopardi.com/posts/gensta... #ElixirLang

Elixir's GenStage Demand (a Visual Explainer)

Concurrency gives us more workers; demand tells us when to stop sending them work. This is what I want to point people to when we talk about "demand".

andrealeopardi.com

The results of the video I posted earlier was this iconic shot 👏 Danny León and his crew smashed it.

Fair play to Danny León, the Spanish Olympic skateboarder. This shot may never be eclipsed.

Generative "AI": The Guitar Hero of Creativity | Discussion

Generative “AI”: The Guitar Hero of Creativity

So, you’re a person who has decided to embark on a creative journey of self-expression — to make art, in other words. First off, this is awesome! The universe needs more people making a…

whatever.scalzi.com

I think with the approach that companies like Oxide seem to be taking, using LLMs not to "go faster" or to skip past having to spend time in understanding a problem, but to instead invest in more exhaustive correctness and validation, absolutely.

Why DNA damage from smoking and UV rays cause cancer in some but not others | Discussion

"Ni siquiera aguantamos una cola en el supermercado o un trayecto de diez minutos en metro sin sacar el teléfono móvil del bolsillo". Miguel Barrero, en El último territorio

El último territorio - Zenda

Nos harán creer de nuevo que responder al instante cualquier mensaje o cualquier llamada es eficiencia en vez de servidumbre.

zendalibros.com

The Silurian Hypothesis (2020) | Discussion

The Silurian Hypothesis by Rich Cohen

January 23, 2020 – Rich Cohen explores the very plausible conspiracy theory that other great civilizations have predated humans on this planet.

theparisreview.org

Kimi K3, Qwen 3.8, and Anthropic's (Potential) Unravelling | Discussion

Kimi K3, Qwen 3.8, and Anthropic's (potential) Unravelling

Kimi K3 and Qwen 3.8 prove open models can reach the frontier. Why the economics of frontier labs favor infrastructure owners — and why Anthropic's model-only position risks unravelling.

emergingtrajectories.com

I see very worrying US companies executives declarations depicting open weight models as a risk. How much sold to the most furious capitalistic vision you need to be, to really believe it is better that a few companies control AI for all the world? They said the same for OSS.

Mozilla: The state of open source AI | Discussion

My notes on Kimi K3, plus some thoughts on what we can still learn from the pelican benchmark even while it becomes further detached from how good the models are at the things that matter (like agentic tool calling across longer conversations) simonwillison.net/2026/Jul/16/...

Kimi K3, and what we can still learn from the pelican benchmark

Chinese AI lab Moonshot AI announced Kimi K3 this morning, describing it as their “most capable model to date, with 2.8 trillion parameters”. It’s currently available via their website and …

simonwillison.net

With AI you can make products no one wants faster than ever!

OSS maintainers who give you software for free are allowed to use any tool they want (and generally run things the way they prefer); unless you're literally paying their salaries, you don't get a say in the matter.

Amazing news! Cervical cancer deaths of 20-24 year olds in England down to ZERO thanks to a vaccine. BBC News - Risk of dying from cervical cancer before 30 'close to zero' after HPV vaccine rollout - BBC News www.bbc.co.uk/news/article... #VaccinesWork

This could happen with your phone. This could happen with your laptop. This could happen with your credit card. It's a long list. Some of us have been working on this, but most large players, public and private, have stayed in fantasyland. Care to join us? www.euractiv.com/news/anthrop...

Anthropic blocks all Mythos access after US order to cut off non-citizens | Euractiv

The company says it disagrees with the US government order and looks to restore access

euractiv.com

We've suspended access to Claude Mythos 5 and Claude Fable 5 | Discussion

We’ve suspended access to Claude Mythos 5 and Claude Fable 5

Claude's Status Page - We’ve suspended access to Claude Mythos 5 and Claude Fable 5.

status.claude.com